Vehicle Collision Avoidance Using Multi-sensor Data Fusion with Dependency information

Bazzi, Youssef A., Rawa Adla, Nizar Al-Holou, and Mohannad Murad

With the advent of sensors technology, the vehicle collision domain has become more attainable. Numerous methods have been discussed in literatures. This paper presents an approach to fuse (integrate) multiple sensor data in order to enhance the assessment of the vehicle collision problem. Our method is based on an implementation of Bayesian probabilistic system with dependency information which proved to produce a more certain, reliable, and robust decision. This method is based on an algorithm to quantify the dependency between these sensors and includes this value in the Bayes’ theorem to estimate the surrounding environment.